This is a phase I, open-label, dose-escalation study to evaluate the preliminary safety, tolerance, pharmacokinetics, and antiviral activity of multiple repeat oral doses of DAPD in treatment-experienced and treatment-naive HIV infected patients. Recent advances in the treatment of patients infected with HIV-1 indicate when viral replication is substantially suppressed, patients can remain clinically stable. This has resulted in HIV-1 infection being treated as a chronic illness requiring long term, multidrug regimens. However, recent reports have described multiple complications of prolonged use of antiretroviral therapy. In addition, loss of control of viral replication in many patients results in the generation of viral strains resistant to multiple antiviral compound. The identification of new anti-HIV-1 agents with improved safety profiles and the ability to suppress the replication of drug-resistant strains of HIV-1 is critical for patients that do not tolerate available therapies or possess mutated viral strains. DAPD, (-)-B-D-2,6-diaminopurine dioxolane, is a nucleoside reverse transcriptase inhibitor that is deaminated in vivo by adenosine deaminase to give (-)-B-D-dioxolane guanine (DXG), producing potent antiviral activity against viral strains that contain mutations that confer resistance to widely used antiretroviral drugs such as 3TC, AZT, d4T. This study is being conducted to test the clinical utility of DAPD when administered to patients infected with HIV-1. Safety, tolerance, pharmacokinetic and pharmacodynamic parameters as well as anti-HIV-1 activity will be evaluated. These effects will be measured in both patients who have detectable viral loads despite receiving antiretroviral therapy, and in patients who are treatment-nanve. A primary goal of this study is to determine if the lack of cross-resistance seen in in vitro testing correlates with antiviral activity of DAPD in the clinical setting. The study will be conducted in 3 parts. In Part A, escalation based upon tolerance and safety at lower doses: 6 treatment-naive patients will receive an initial oral dose of DAPD and pharmacokinetic sampling will ensue over the next 12 hours prior to receiving the second dose. Dosing will be done on a BID schedule except for Day 15, when only a single dose will be administered in order to obtain 24- and 48-hour post-treatment PK samples. After all 6 patients in a cohort have completed 14 days of treatment at the preceding dose level, sponsor will review safety data, including adverse events and laboratory safety test data. Escalation to the next highest dose may occur if less than or equal to 2/6 patients in the dose cohort have experienced treatment-limiting clinical adverse events that led to the discontinuation of study drug and was considered by both the investigator and sponsor to be probably related to the study drug. The same dose escalation procedure will apply to all subsequent dose escalations. In Part B, the BID dose(s) that produces a satisfactory effect on HIV viral load and demonstrates acceptable safety and tolerance will be tested in a cohort of 6 to 12 treatment-experienced patients that have discontinued prior anti-retroviral treatment for a minimum of 7 days. Two groups of treatment-experienced patients will be enrolled in Part B of the study after prospective genotypic analysis indicates that the patient does not harbor HIV containing the following mutations: K65R, L74V, or 69 SSS insert. In Part C, the BID dose(s) that produce a satisfactory antiviral response may be tested as a total daily dose on a QD schedule in another cohort of patients. Anti-viral response as measured by HIV-1 RNA PCR will be assessed at Screen, Baseline (prior to the first dose of study medication), Days 1, 2, 3, 4, 5, 8, 10, 12, and predose and at 12, 24, and 48 hours after the last dose on Day 15 or premature termination.